TABLE 2.

Summary risk estimates of the association between breastfeeding and ovarian cancer risk

	Studies	Summary RR (95% CI)	Q statistic	I2	Ph1	Ph2
	n			%
Overall	32	0.76 (0.69–0.83)	69.40	55.3	<0.001	—
Subgroup analyses
Study design						0.090
Cohort studies	5	0.88 (0.78, 0.99)	0.73	0	0.947
Case-control studies	27	0.74 (0.67, 0.82)	62.36	58.3	<0.001
Exposure assessment						0.065
Trained interviewer	15	0.68 (0.57, 0.80)	51.59	72.9	<0.001
Self-administered questionnaire	12	0.82 (0.75, 0.90)	7.06	0	0.794
Type of control subjects						0.158
Population based	16	0.73 (0.68, 0.78)	22.91	34.5	0.086
Hospital based	10	0.78 (0.60, 1.02)	31.01	71.0	<0.001
Study population						0.862
Asians	7	0.69 (0.53, 0.89)	3.99	0	0.678
Americans	13	0.71 (0.63, 0.81)	35.90	66.6	<0.001
Europeans	8	0.85 (0.69, 1.06)	19.75	64.6	0.006
Cancer grading						0.645
Invasive	5	0.62 (0.53, 0.72)	6.14	34.9	0.189
Borderline	4	0.57 (0.44, 0.74)	2.53	0	0.470
Cancer histotype						0.267
Serous	7	0.82 (0.68, 0.99)	13.91	56.9	0.031
Mucinous	6	0.80 (0.64, 1.00)	7.10	29.6	0.213
Endometrioid	3	0.65 (0.47, 0.89)	2.10	5.0	0.349
Clear cell	2	0.67 (0.39, 1.15)	0.92	0	0.336
Adjustment for confounders
Parity						0.285
Yes	22	0.78 (0.71, 0.85)	42.23	50.3	0.004
No	10	0.70 (0.57, 0.87)	18.55	51.5	0.029
BMI						0.803
Yes	5	0.79 (0.69, 0.91)	4.43	9.7	0.351
No	27	0.75 (0.68, 0.83)	64.71	59.8	<0.001
OC3 use						0.782
Yes	17	0.77 (0.70, 0.84)	32.56	50.9	0.008
No	15	0.87 (0.69, 1.09)	34.97	60.6	0.001
Smoking						0.505
Yes	7	0.71 (0.57, 0.88)	15.32	60.8	0.018
No	25	0.77 (0.70, 0.85)	54.00	55.6	<0.001

¹P value for heterogeneity within each subgroup.

²P value for heterogeneity between subgroups with meta-regression analysis.

³OC, oral contraceptive.